EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH (CERN) CERN-PH-EP/2011-187 2018/07/30 CMS-FWD-10-005 Exclusive gg m+m production in proton-proton ! − collisions at ps = 7 TeV The CMS Collaboration∗ Abstract A measurement of the exclusive two-photon production of muon pairs in proton- + proton collisions at ps = 7 TeV, pp pm m−p, is reported using data correspond- ! 1 ing to an integrated luminosity of 40 pb− . For muon pairs with invariant mass greater than 11.5 GeV, transverse momentum pT(m) > 4 GeV and pseudorapidity h(m) < 2.1, a fit to the dimuon p (m+m ) distribution results in a measured cross j j T − section of s(p pm+m p) = 3.38+0.58 (stat.) 0.16 (syst.) 0.14 (lumi.) pb, con- ! − 0.55 ± ± sistent with the theoretical prediction− evaluated with the event generator LPAIR. The ratio to the predicted cross section is 0.83+0.14 (stat.) 0.04 (syst.) 0.03 (lumi.). The 0.13 ± ± characteristic distributions of the muon− pairs produced via gg fusion, such as the muon acoplanarity, the muon pair invariant mass and transverse momentum agree with those from the theory. Submitted to the Journal of High Energy Physics arXiv:1111.5536v2 [hep-ex] 11 Jan 2012 ∗See Appendix A for the list of collaboration members 1 1 Introduction The exclusive two-photon production of lepton pairs may be reliably calculated within the framework of quantum electrodynamics (QED) [1] (Fig. 1), within uncertainties of less than 1% associated with the proton form factor [2]. Indeed, detailed theoretical studies have shown that corrections due to hadronic interactions between the elastically scattered protons are well below 1% and can be safely neglected [3]. The unique features of this process, like the extremely small pair transverse momentum and acoplanarity (defined as 1 Df(m+m )/p ), stem from − j − j the very small virtualities of the exchanged photons. At the Tevatron, the exclusive two-photon production of electron [4, 5] and muon [5, 6] pairs in pp collisions has been measured with the CDF detector. Observations have been made of QED signals, leading to measurements of exclusive charmonium photoproduction [6] and searches for anomalous high-mass exclusive dilepton production [5]. However, all such measurements have very limited numbers of selected events because the data samples were restricted to sin- gle interaction bunch crossings. The higher energies and increased luminosity available at the Large Hadron Collider (LHC) will allow significant improvements in these measurements, if this limitation can be avoided. As a result of the small theoretical uncertainties and character- istic kinematic distributions in gg m+m , this process has been proposed as a candidate for ! − a complementary absolute calibration of the luminosity of pp collisions [1–3]. Unless both outgoing protons are detected, the semi-exclusive two-photon production, involv- ing single or double proton dissociation (Fig. 1, middle and right panels), becomes an irre- ducible background that has to be subtracted. The proton-dissociation process is less well determined theoretically, and in particular requires significant corrections due to proton rescat- tering. This effect occurs when there are strong-interaction exchanges between the protons, in addition to the two-photon interaction. These extra contributions may alter the kinematic dis- tributions of the final-state muons, and may also produce additional low-momentum hadrons. As a result, the proton-dissociation process has significantly different kinematic distributions compared to the pure exclusive case, allowing an effective separation of the signal from this background. p p p p + + µ µ+ µ γ γ γ γ γ γ µ− µ− µ− p p p p p Figure 1: Schematic diagrams for the exclusive and semi-exclusive two-photon production of muon pairs in pp collisions for the elastic (left), single dissociative (center), and double dissociative (right) cases. The three lines in the final state of the center and right plots indicate dissociation of the proton into a low-mass system N. In this paper, we report a measurement of dimuon exclusive production in pp collisions at ps = 7 TeV for the invariant mass of the pair above 11.5 GeV, with each muon having trans- verse momentum p (m) > 4 GeV and pseudorapidity h(m) < 2.1 (where h is defined as T j j ln(tan(q/2))). This measurement is based on data collected by the Compact Muon Solenoid − (CMS) experiment during the 2010 LHC run, including beam collisions with multiple interac- tions in the same bunch crossing (event pileup), and corresponding to an integrated luminosity 1 of 40 pb− with a relative uncertainty of 4% [7]. 2 3 Simulated Samples The paper is organized as follows. In Section 2, a brief description of the CMS detector is provided. Section 3 describes the data and samples of simulated events used in the analysis. Section 4 documents the criteria used to select events, and Section 5 the method used to extract the signal yield from the data. The systematic uncertainties and cross-checks performed are discussed in Section 6, while Section 7 contains plots comparing the selected events in data and simulation. Finally, the results of the measurement are given in Section 8 and summarized in Section 9. 2 The CMS detector A detailed description of the CMS experiment can be found elsewhere [8]. The central fea- ture of the CMS apparatus is a superconducting solenoid, of 6 m internal diameter. Within the field volume are the silicon pixel and strip tracker, the crystal electromagnetic calorime- ter, and the brass/scintillator hadronic calorimeter. Muons are measured in gaseous detectors embedded in the iron return yoke. Besides the barrel and endcap detectors, CMS has exten- sive forward calorimetry. CMS uses a right-handed coordinate system, with the origin at the nominal collision point, the x axis pointing to the center of the LHC ring, the y axis pointing up (perpendicular to the plane of the LHC ring), and the z axis along the anticlockwise-beam direction. The azimuthal angle f is measured in the x-y plane. Muons are measured in the window h < 2.4, with detection planes made using three systems: drift tubes, cathode strip j j chambers, and resistive plate chambers. Thanks to the strong magnetic field, 3.8 T, and to the high granularity of the silicon tracker (three layers consisting of 66 million 100 150 mm2 pix- × els followed by ten microstrip layers, with strips of pitch between 80 and 180 mm), the pT of the muons matched to silicon tracks is measured with a resolution better than 1.5%, for p ∼ T less than 100 GeV. The first level of the CMS trigger system, composed of custom hardware processors, uses information from the calorimeters and muon detectors to select (in less than 1 ms) the most interesting events. The High Level Trigger processor farm further decreases the event rate from 50–100 kHz to a few hundred Hz, before data storage. 3 Simulated Samples The LPAIR 4.0 event generator [9, 10] is used to produce simulated samples of two-photon production of muon pairs. The generator uses full leading-order QED matrix elements, and the cross sections for the exclusive events depend on the proton electromagnetic form-factors to account for the distribution of charge within the proton. For proton dissociation, the cross sections depend on the proton structure function. In order to simulate the fragmentation of the dissociated proton into a low-mass system N, the LUND model shower routine [11] imple- mented in the JETSET software [12] is used with two different structure functions. For masses of 2 2 the dissociating system mN < 2 GeV and photon virtualities Q < 5 GeV , the Brasse “cluster” fragmentation is chosen [13], while for the other cases the Suri-Yenni “string” fragmentation is applied [14]. In the first case, the low-mass system N mostly decays to a D+ or D++ reso- nance, which results in low-multiplicity states. In the second case, the high-mass system usu- ally decays to a variety of resonances (D, r, W, h, K), which produce a large number of forward protons, pions, neutrons and photons. No corrections are applied to account for rescattering effects. In general, these effects are expected to increase with the transverse momentum of the 2 + muon pair, modifying the slope of the pT(m m−) distribution [3]. The inclusive Drell–Yan (DY) and quantum chromodynamic (QCD) dimuon backgrounds are simulated with PYTHIA v. 6.422 [15], using the Z2 underlying event tune [16]. All these samples 3 are then passed through the full GEANT4 detector simulation [17] in order to determine the signal and background efficiencies after all selection criteria are applied. 4 Event selection The analysis uses a sample of pp collisions at ps = 7 TeV, collected during 2010 at the LHC 1 1 and corresponding to an integrated luminosity of 40 pb− . The sample includes 36 pb− of data 1 passing the standard CMS quality criteria for all detector subsystems, and 4 pb− in which the quality criteria are satisfied for the tracking and muon systems used in the analysis. From the sample of triggered events, the presence of two reconstructed muons is required. Then the exclusivity selection is performed to keep only events with a vertex having no tracks other than those from the two muons. Finally, the signal muons are required to satisfy identification criteria, and kinematic constraints are imposed using their four-momentum. All selection steps are described in the following sections. 4.1 Trigger and muon reconstruction Events are selected online by triggers requiring the presence of two muons with a minimum pT of 3 GeV.